Allele specific expression in worker reproduction genes in the ...

A peer-reviewed version of this preprint was published in PeerJ on 14 July 2015. View the peer-reviewed version (peerj.com/articles/1079), which is the preferred citable publication unless you specifically need to cite this preprint. Amarasinghe HE, Toghill BJ, Nathanael D, Mallon EB. (2015) Allele specific expression in worker reproduction genes in the bumblebee Bombus terrestris. PeerJ 3:e1079 https://doi.org/10.7717/peerj.1079

Allele specific expression in worker reproduction genes in the bumblebee Bombus terrestris Harindra E. Amarasinghe1,* , Bradley J. Toghill2,* , and Eamonn B. Mallon1 1 Department

of Biology, University of Leicester of Cardiovascular Sciences, University of Leicester * Joint first authors 2 Department

PrePrints

ABSTRACT Genomic imprinting is the differential expression of alleles, with the expression being dependent upon the sex of the parent from which it was inherited. Hymenopteran insects (ants, bees and wasps) are emerging as potential models for genomic imprinting and epigenetics. As a first step in establishing the possibility of genomic imprinting in the bumblebee, Bombus terrestris, we search for allele specific expression in twelve genes associated with worker reproduction. We found that the patrigene (allele from the father) is more expressed than the matrigene (allele from the mother) in Ecdysone 20 monooxygenase. This enzyme catalyses the reaction which turns the ecdysteroid ecdysone into 20-hydroxyecdysone, also an ecdysteroid. Both of these ecdysteroids are important for worker reproduction in the bumblebee.

Keywords:

Social insect, Hymenoptera, Genomic imprinting, Ecdysone

INTRODUCTION Genomic imprinting is the differential expression of alleles, with the expression being dependent upon the sex of the parent from which it was inherited (Haig, 2000). Genomic imprinting is an important area of research in evolutionary biology (Hurst, 1997), in plant breeding (Gehring, 2013) and has relevance to some human cancers and developmental syndromes (Lee and Bartolomei, 2013). Several recent reviews have heralded hymenopteran insects (ants, bees and wasps) as important emerging models for genomic imprinting and epigenetics in general (Weiner and Toth, 2012; Yan et al., 2014; Welch and Lister, 2014; Glastad et al., 2011). This is based mainly on data showing a fundamental role for methylation in their biology (Chittka et al., 2012). Methylation, the addition of a methyl group to a cytosine, is an epigenetic marker of genomic imprinting in mammals and flowering plants (Glastad et al., 2011). In a recent paper (Amarasinghe et al., 2014), we showed that methylation is important in worker reproduction in the bumblebee, Bombus terrestris. We found methylation differences between the genomes of queenless reproductive workers and queenless non-reproductive workers. In a follow up experiment, queenless workers whose genomes had experimentally altered methylation (fed 5-aza-2’-deoxycytidine) were more aggressive and more likely to develop ovaries compared with control queenless workers. This is interesting as Haig’s kinship theory, the leading explanation for the evolution of genomic imprinting (Patten et al., 2014), suggests that the genes associated with bumblebee worker reproduction could be imprinted (Queller, 2003). As a next step in establishing the possibility of genomic imprinting in bumblebees, we searched for allele specific expression in worker reproduction genes. Previous work has found allele specific expression in a number of loci in the ants Camponotus floridanus and Harpegnathos saltator (Bonasio et al., 2012). Based on Haig’s theory we chose twelve genes previously associated with worker reproduction in bees (see Table 1). We looked for polymorphisms in their exonic DNA in queens and their daughter workers using single strand conformation polymorphism (SSCP) and sanger sequencing. SSCP relies on the principle that the electrophoretic mobility of a single-stranded DNA molecule is dependent on its structure (nucleotide sequence) and size. In the absence of the complementary strand, DNA becomes unstable and reanneals to itself to form conformations; hairpins, pseudoknots and triple

PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

helices (Nielsen et al., 1995). These conformations vary according to the primary sequence of the molecule, such that a single nucleotide difference in DNA could dramatically affect the strand’s mobility through a gel due to its unique 3D structure. If we found that the workers possess an allele from the queen and another allele not present in the queen, this allele must be from the father. That is we have identified a matrigene (allele from the mother) and patrigene (allele from the father) at this locus. If we found this, we carried out an allele specific qPCR to ascertain if this locus displayed allele specific expression.

PrePrints

METHODS Identification of candidate genes and designing primers Twelve social insect genes previously associated with differential expression in queens, reproducing workers and non-reproducing workers were selected via a literature search (Table 1). Sequences for all selected candidate genes were obtained from Apis mellifera genome data, available in NCBI. Apis mellifera data was BLASTed against the Bombus terrestris Nucleotide library (NCBI) in order to find the homolog in Bombus terrestris. Primers were designed to the exonic regions using Geneious Pro (version 5.5.6) and primer 3 version 0.4.0 (http://frodo.wi.mit.edu). The focus was on exonic regions to ensure that the same loci was present in the cDNA for the allele specific qPCR analysis. Apis mellifera Chymotrypsin Gemini Cabut Ecdysone 20 monooxygenase Yolkless Epidermal growth factor receptor Ribosomal Protein L26

Odorant receptor2 Dop3 D2-like dopamine receptor Megator

Bombus terrestris Chymotrypsin-1-like (LOC100648122) Upstream binding protein 1 -like (LOC100650338) Zinc finger protein 691-like (LOC100642767) Ecdysone 20 monooxygenase -like (LOC100649449) Vitellogenin receptor-like (LOC100649042) Epidermal growth factor receptor like (LOC100645521) Ribosomal Protein L26 like (LOC100648461) Or2 odorant receptor 2/ Queen mandibular pheromone (QMP) co-receptor (LOC100631089) D2 like dopamine receptor (LOC100644210) Megator TPR like nucleoprotein (LOC100645723)

Biological function Upregulated in bumblebee nonreproductive workers (Pereboom et al., 2005) Upregulated in honeybee reproductive workers (Jarosch et al., 2011) Upregulated in honeybee nonreproductive workers (Cardoen et al., 2011) Upregulated in honeybee reproductive workers (Cardoen et al., 2011) Upregulated in honeybee reproductive workers (Cardoen et al., 2011) Upregulation of EGFR initiates ovary activation in queenless honeybee workers (Formesyn et al., 2014) Differentially expressed in honeybee reproductive and non-reproductive workers (Thompson et al., 2007) Upregulated in honeybee sterile workers (Grozinger et al., 2007) Upregulated in honeybee nonreproductive workers (Vergoz et al., 2012) Upregulated in honeybee reproductive workers (Cardoen et al., 2011)

Ecdysteroid regulated gene E93/Mblk-1 transcription factor

Mushroom body large-type Kenyon cell specific protein 1 -like (LOC100645656)

Upregulated in honeybee reproductive workers (Cardoen,et al., 2011)

Ecdysone inducible gene L2/ ImpL2

Neural/ectodermal development factor IMP-L2-like (LOC100645498)

Upregulated in honeybee nonreproductive workers (Cardoen,et al., 2011)

Table 1. Candidate genes selected from the literature search. NCBI gene ID is given within parenthesis. 2/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

PrePrints

Samples The queen and 5 randomly selected workers from each colony were used for SSCP analysis. Most candidate genes were tested in four different bumblebee colonies. Chymotrypsin, Gemini, Cabut and Yolkless were tested in colonies 1-4. Another four colonies (5-8) were used to test Epidermal growth factor receptor, Ribosomal protein L26, Odorant receptor 2, Dop3, Megator, Ecdysteroid regulated gene E93 (Mblk1) and Ecdysone inducible gene L2 (IMP-L2). Ecdysone 20 monooxygenase-like was tested in eight colonies. All qPCR data are based on bees from colony 5. DNA and RNA extraction and cDNA synthesis Bees were frozen in liquid nitrogen and then stored at -80o C. Genomic DNA for SSCP analysis was extracted from each queen and respective worker bees using the Qiagen DNA Micro kit according to manufacturer’s instructions. Concentration of total genomic DNA was measured using the NanoDrop 1000 Spectrophotometer. A 30mg sample of frozen tissue was ground with mortar and pestle on dry ice. RNA was extracted with the QIAGEN RNeasy Mini Kit according to manufacturer’s instructions. Any DNA contamination present in the above RNA extractions was removed according to Amplification Grade DNase I Kit protocol (Sigma-Aldrich), prior to the synthesis of cDNA. Concentrations of DNase treated RNA was determined by the NanoDrop 1000 spectrophotometer. cDNA was synthesized from a 8µl sample of RNA using the Tetro cDNA synthesis Kit (Bioline) as per manufacturer’s instructions. Synthesized cDNA was stored at -80o C. PCR amplifications For each primer set, a 25µl reaction volume (60ng of DNA, 12.5µl YB-Taq 2x Buffer, 1.5µl of each forward and reverse primer (10µlM/µl), 1µl of 10mM MgCl2 and 6.5µl of dH2 0) was run using the following conditions: an initial denaturation for 5 min at 94o C, 30 cycles of 30s at 94o C, 30s each at the relevant annealing temperature followed by a final extension of 10 minutes at the relevant extension temperature and a holding step of 4o C. The sequences and annealing and extension temperatures used for each primer set are in supplementary table S1. Prior to SSCP analysis, each PCR product (10µl) was checked on a 3% agarose gel. If the correct size of amplicon was obtained, then the rest of the sample (15µl) was used for SSCP. SSCP analysis SSCP analysis was carried out according to Gasser et al. (2007) using GMA wide mini S-2x25 gels (Elchrom scientific). Sample denaturing solution was prepared by mixing 990µl of 95% formamide with 10µl of 1M NaOH just prior to use. 4µl of the PCR product was denatured with 7µl of denaturing mixture, incubated in a thermocycler at 94o C for 10 minutes and immediately chilled on ice for 5 minutes. The temperature of the running buffer (1x TAE) in the Origins gel tank (Elchrom scientific) was kept constant at 9o C. 7µl of the denatured PCR product was mixed with 2µl of Elchrom loading dye and loaded in to a well on the gel. The gels were run at 72V. The electrophoretic running times were varied depending on the fragment size; 10 hours for 150 - 200bp fragment length, 12 hours for 200 - 250bp fragment length, 15 hours for 250 - 350bp fragment length and 17 hours for 350 - 450bp fragment length. Following electrophoresis, the gels were stained for 30 minutes with SybrGold (Invitrogen) (1:10000 diluted in TAE) and destained with 100ml of 1xTAE buffer for a further 30 minutes. If a polymorphic banding pattern among the queen and her 5 workers was observed during SSCP, another SSCP was run to confirm the reproducibility of those results. The genomic DNA of those queen and worker bees were amplified with their respective primers (see supplementary table 1) and PCR products were sent for commercial clean up and sanger sequencing. All sequencing results were blasted against NCBI, Bombus terrestris nucleotide library to verify if the correct sequence was amplified. Sequencing results were analyzed using the heterozygote analysis module in Geneious version 7.3.0 to identify heterozygotic nucleotide positions. Allele specific PCR Allele specific PCR was used to confirm the maternal and paternal alleles identified during heterozygote analysis. Allele specific primers were designed using Batch primer 3 program (http://probes.pw. usda.gov/batchprimer3/) to cover the heterozygotic nucleotide positions identified above. Two forward primers specific to either maternal (F1) or paternal (F2) allele sequences and a common reverse 3/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

PrePrints

primer were designed. Genomic DNA of the queen and 5 heterozygous workers in each colony, were PCR amplified with these allele specific primers (Table 2). PCR products were checked on a 3% agarose gel. When using allele specific primers, only the allele which includes the relevant snp would be amplified. Primers which amplified the snp region successfully were used for qPCR analysis. Allele specific quantitative PCR Reference primers were designed according to Gineikiene et al. (2009). A common forward primer was designed to the same target heterozygote sequence, upstream of the heterozygote nucleotide position, leaving the same common reverse primer previously used with allele specific primers (see reference sequences in Table 2). The reference primers measure the total expression of the gene, whereas the allele specific primers measure the amount of expression due to the allele. Thus the expression difference between the reference and allele specific primers would be the relative expression due to the allele. Each heterozygous locus was run for 3 different reactions; maternal (F1), paternal (F2) and reference (Table 2). Three replicate samples were run for each reaction. All reactions were prepared by the Corbett robotics machine, in 96 well qPCR plates (Thermo Scientific, UK). The qPCR reaction mix (20µl) was composed of 1µl of diluted cDNA (50ng/µl), 1µl of forward and reverse primer (5µM/µl each, Table 2), 10µl 2X SYBR Green JumpStart Taq ReadyMix (Sigma Aldrich, UK) and 7µl ddH2 0. Samples were run in a PTC-200 MJ thermocycler. The qPCR profile was; 4 minutes at 95o C denaturation followed by 40 cycles of 30s at 95o C, 30s at the relevant annealing temperature (Table 2) and 30s at 72o C and a final extension of 5 minutes at 72o C. Gene

Heterozygote position

Ecdysone 20 monooxygenase like

48 (A/G)

Ecdysone 20 monooxygenase like internal reference IMP-L2-like IMP-L2-like internal reference

Primer sequence (5’-3’) F1: GCGGAAGCCGTCAGG F2: TTAGCGGAAGCCGTCAGA R: GCGAGGCCGTAAAGTGTAT F: GATTTAGCGGAAGCCGTCAG R: GCGAGGCCGTAAAGTGTAT

253 (A/G)

F1: ACTTGCCAAGCCAAGTCTG F2: CACTTGCCAAGCCAAGTCTA R: TTCGAGCCACTTCCTTTTCG F: CTACACTTGCCAAGCCAAGTCT R: TTCGAGCCACTTCCTTTTCG

TA (o C)

Product size (bp)

58

34

59

36

59.5

205

59.5

207

Table 2. Allele specific primers used for gene expression analysis. F1 = Forward primer 1, F2 = Forward primer 2, R= Common reverse primer. The snp present is located at the 3’ end of each forward primer (marked in red). TA = Annealing temperature.

Data analysis Median Ct was calculated for each set of three technical replicates. A measure of relative expression (∆Ct ) was calculated for each parental allele in each worker bee as follows:

∆Ct (maternal) = Ct(re f erence) −Ct (maternal)

(1)

∆Ct (paternal) = Ct(re f erence) −Ct (paternal)

(2)

Matched paired t-tests was performed to check if the allele specific expression values are significantly different among the two parental alleles. All statistical analysis was carried out using R (3.1.0) (Team, 2013). 4/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

RESULTS

PrePrints

SSCP analysis Exon coverage for each gene is given in Table 3. We found no polymorphisms in nine genes out of the twelve candidate genes tested (see Table 3). Figure 1 shows representative examples of these gels with the queen and her workers sharing the same banding pattern. Gene name Chymotrypsin-1-like Upstream-binding protein 1-like Zinc finger protein 691-like Vitellogenin receptor-like Epidermal growth factor receptor like 60S Ribosomal Protein L26 like Or2 odorant receptor 2 D2 like dopamine receptor Megator TPR like nucleoprotein Ecdysone 20-monooxygenase-like Mushroom body large-type Kenyon cell-specific protein 1-like Neural/ectodermal development factor IMP-L2-like

Coverage (%) 92 93 70 30 21 32 25 54 17 37 35 47

Polymorphism Absent Absent Absent Absent Absent Absent Absent Absent Absent Present Present Present

Table 3. SSCP exon coverage. Percentage exon coverage for each gene was calculated as the sum of all tested amplicon lengths as a fraction of the total length of mRNA per gene.

Figure 1. SSCP gel results of six genes (a - f) with no queen-worker variations (homozygous banding patterns). The queen (Q) and 5 workers (W1-W5) are represented in each colony. Compared with the queen, workers in colony 3, 5 and 7 showed a heterozygous banding pattern in 3 genes; ecdysone 20-monooxygenase-like (ecdysone 20-monooxygenase-like), IMPL-2-like and MBLK1-like (Figure 2). Sequences of polymorphic loci We sequenced the three loci showing heterozygous banding patterns in SSCP gels. In ecdysone 20monooxygenase-like, the queen sequence is homozygous (Figure 2). At the snp (2474th base pair of LOC100649449) the queen has a guanine (G), while all of her workers show double peaks corresponding to both guanine (G) and adenine (A) bases (see supplementary table S2 for sequences). From this we 5/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

PrePrints

Figure 2. SSCP allelic polymorphism in IMPL2-like, Mblk1-like and Ecdysone 20-monooxygenase-like.

identified the matrigene as containing guanine and the patrigene as containing adenine. A similar SSCP banding pattern was found for IMP-L2-like (Figure 2) Again the queen had a guanine where as workers contained a guanine and an adenine, this time at the 5130th base pair of LOC100645498 (supplementary table S2)). Mblk1-like has several snps in the amplified region (Figure 2 and supplementary table S2)). Allele specific PCR Allele specific primers designed for ecdysone 20-monooxygenase-like and IMP-L2-like worked successfully with genomic DNA and cDNA to produce the expected fragment lengths. They were used for gene expression analysis in the next section. Amplification of MBLK-1 using allele specific primer sets was unsuccessful possibly due to the large number of snps. Thus, we did not continue with qPCR analysis for MBLK1-like. Allele specific qPCR The patrigene showed significantly increased expression compared to the matrigene in ecdysone 20monooxygenase-like (Figure 3) (paired t-test:t = -4.4593, df = 4, p = 0.01117). For IMP-L2-like, there was no significant difference in expression between the parental alleles (Figure 4) (paired t-test: t = -1.5572, df = 4, p = 0.1944).

DISCUSSION Using a candidate gene approach we found evidence for allele specific expression in the bumblebee, Bombus terrestris. Out of twelve genes examined during this study, we found exonic variation in only three genes; MBLK1-like, IMP-L2-like and ecdysone 20-monooxygenase-like. Of these we were able to carry out allele specific qPCR on IMP-L2-like and ecdysone 20-monooxygenase-like. We found allele specific expression in ecdysone 20-monooxygenase-like but not IMP-L2-like. Use of SSCP to find exonic variation is challenging. SSCP detects variation in fragments up to 500bp size with a high resolution of 1bp. However, the sensitivity of SSCP decreases when the fragment length exceed 200bp (Weber et al., 2005). Thus medium length fragments around 200bp were used for this analysis. Covering the full exome using SSCP would be a time consuming and labour intensive process. Added to this, variation in protein coding exons is expected to be rarer than in introns (Castle, 2011). One possiblility would be to look at the exons in untranslated regions (UTRs), which would be expected to be more variable than protein coding exons(Araujo et al., 2012; Lytle et al., 2007). Our expression analysis used the bees’ whole bodies. Therefore gene expression patterns observed during this analysis should represent the overall expression of all body tissues. However, potentially it means allele specific expression which is only found in some tissues would be masked by the overall response. We found allele specific expression in Ecdysone 20 monooxygenase. Ecdysone 20 monooxygenase catalyses the reaction which turns ecdysteroid ecdysone into 20-hydroxyecdysone, also an ecdysteroid. 6/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

Rawdata

Difference ●

●

−5.0

●

1.0

●

●

−6.0

deltaCt

●

● ●

●

PrePrints

●

● ●

0.4

−7.0

−6.5

●

0.8

●

0.6

●

Difference:paternal−maternal

−5.5

●

●

●

Maternal

Paternal Allele

Median+/−95%CI

Figure 3. Expression (measured as ∆Ct ) differences between maternal and paternal alleles of ecdysone 20-monooxygenase-like. The first plot shows the expression data as individual dots. The diagonal lines join data from the same bee. Boxplots represent the distributions. The second plot shows the difference between paternal and maternal expression as individual dots. The red vertical bar represents the 95% confidence interval that these data are difference from zero based on a one-sample wilcoxon test. An up-regulation of ecdysone 20-monooxygenase-like was observed in egg laying honeybee workers compared to non-reproductive workers (Cardoen et al., 2011). Generally, ecdysteroids have been identified as key regulators of B.terrestris worker reproduction (Geva et al., 2005). Ecdysteriods are key compounds involved in ovary activation, regulating agonistic behaviour and establishing the dominance hierarchy in workers and queens (Geva et al., 2005). Our analysis found allele specific expression, but allele specific expression is known to be caused by a number of other processes besides genomic imprinting (Palacios et al., 2009). Given this it is still interesting to note that ecdysone 20-monooxygenase-like in its worker reproduction role is exactly the type of gene that is predicted to be imprinted according to Haig’s kinship theory for the evolution of genomic imprinting (Queller, 2003). Clearly the candidate gene approach is limited in its application. Next generation sequencing technology allows gene expression analysis at genome-wide scale (RNA-seq). Several recent papers have applied RNA-seq to search unbiasedly for novel imprinted genes in mammals (Okae et al., 2012; DeVeale et al., 2012; Gregg et al., 2010; Wang and Clark, 2014) and flowering plants (Gehring, 2013). Our results suggest that this could be a potentially fruitful avenue for research in the social insects.

ACKNOWLEDGMENTS Thanks to Sally Adams for discussions.

REFERENCES Amarasinghe, H. E., Clayton, C. I., and Mallon, E. B. (2014). Methylation and worker reproduction in the bumble-bee (Bombus terrestris). Proceedings of the Royal Society B: Biological Sciences, 281(1780):20132502. Araujo, P. R., Yoon, K., Ko, D., Smith, A. D., Qiao, M., Suresh, U., Burns, S. C., and Penalva, L. O. F. (2012). Before It Gets Started: Regulating Translation at the 5’ UTR. International Journal of Genomics, 2012:e475731. Bonasio, R., Li, Q., Lian, J., Mutti, N. S., Jin, L., Zhao, H., Zhang, P., Wen, P., Xiang, H., Ding, Y., Jin, Z., Shen, S. S., Wang, Z., Wang, W., Wang, J., Berger, S. L., Liebig, J., Zhang, G., and Reinberg, D. (2012). Genome-wide and Caste-Specific DNA Methylomes of the Ants Camponotus floridanus and Harpegnathos saltator. Current Biology, 22(19):1755–1764. 7/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

Difference 1.5

Rawdata

●

●

●●

−1.5

●

1.0

●

●

●

PrePrints

●

●

−0.5

−3.0

−2.5

●

●

0.5

●

0.0

●

−2.0

deltaCt

●

Difference:paternal−maternal

●

●

●

●

Maternal

Paternal Allele

Median+/−95%CI

Figure 4. Expression (measured as ∆Ct ) differences between maternal and paternal alleles of Ecdysone-inducible gene L2 (IMP-L2-like). The first plot shows the expression data as individual dots. The diagonal lines join data from the same bee. Boxplots represent the distributions. The second plot shows the difference between paternal and maternal expression as individual dots. The red vertical bar represents the 95% confidence interval that these data are difference from zero based on a one-sample wilcoxon test. The horizontal line represents zero.

Cardoen, D., Wenseleers, T., Ernst, U. R., Danneels, E. L., Laget, D., DE Graaf, D. C., Schoofs, L., and Verleyen, P. (2011). Genome-wide analysis of alternative reproductive phenotypes in honeybee workers. Molecular Ecology, 20(19):4070–4084. Castle, J. C. (2011). SNPs Occur in Regions with Less Genomic Sequence Conservation. PLoS ONE, 6(6):e20660. Chittka, A., Wurm, Y., and Chittka, L. (2012). Epigenetics: The making of ant castes. Current Biology, 22(19):R835–R838. DeVeale, B., van der Kooy, D., and Babak, T. (2012). Critical Evaluation of Imprinted Gene Expression by RNA–Seq: A New Perspective. PLoS Genet, 8(3):e1002600. Gasser, R. B., Hu, M., Chilton, N. B., Campbell, B. E., Jex, A. J., Otranto, D., Cafarchia, C., Beveridge, I., and Zhu, X. (2007). Single-strand conformation polymorphism (SSCP) for the analysis of genetic variation. Nature Protocols, 1(6):3121–3128. Gehring, M. (2013). Genomic Imprinting: Insights From Plants. Annual Review of Genetics, 47(1):187– 208. Geva, S., Hartfelder, K., and Bloch, G. (2005). Reproductive division of labor, dominance, and ecdysteroid levels in hemolymph and ovary of the bumble bee Bombus terrestris. Journal of Insect Physiology, 51(7):811–823. Gineikiene, E., Stoskus, M., and Griskevicius, L. (2009). Single Nucleotide Polymorphism-Based System Improves the Applicability of Quantitative PCR for Chimerism Monitoring. The Journal of Molecular Diagnostics : JMD, 11(1):66–74. Glastad, K. M., Hunt, B. G., Yi, S. V., and Goodisman, M. a. D. (2011). DNA methylation in insects: on the brink of the epigenomic era. Insect Molecular Biology, 20(5):553–565. Gregg, C., Zhang, J., Weissbourd, B., Luo, S., Schroth, G. P., Haig, D., and Dulac, C. (2010). Highresolution analysis of parent-of-origin allelic expression in the mouse brain. Science (New York, N.Y.), 329(5992):643–648. Haig, D. (2000). The kinship theory of genomic imprinting. Annual Review of Ecology and Systematics, 31:9–32. Hurst, L. D. (1997). Evolutionary theories of genomic imprinting. In Reik, W. and Surani, A., editors, Genomic imprinting. IRL Press, Oxford. 8/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015

PrePrints

Lee, J. and Bartolomei, M. (2013). X-Inactivation, Imprinting, and Long Noncoding RNAs in Health and Disease. Cell, 152(6):1308–1323. Lytle, J. R., Yario, T. A., and Steitz, J. A. (2007). Target mRNAs are repressed as efficiently by microRNAbinding sites in the 5’ UTR as in the 3’ UTR. Proceedings of the National Academy of Sciences of the United States of America, 104(23):9667–9672. Nielsen, D. A., Novoradovsky, A., and Goldman, D. (1995). SSCP primer design based on single-strand DNA structure predicted by a DNA folding program. Nucleic Acids Research, 23(12):2287–2291. Okae, H., Hiura, H., Nishida, Y., Funayama, R., Tanaka, S., Chiba, H., Yaegashi, N., Nakayama, K., Sasaki, H., and Arima, T. (2012). Re-investigation and RNA sequencing-based identification of genes with placenta-specific imprinted expression. Human Molecular Genetics, 21(3):548–558. Palacios, R., Gazave, E., Go˜ni, J., Piedrafita, G., Fernando, O., Navarro, A., and Villoslada, P. (2009). Allele-Specific Gene Expression Is Widespread Across the Genome and Biological Processes. PLoS ONE, 4(1):e4150. Patten, M. M., Ross, L., Curley, J. P., Queller, D. C., Bonduriansky, R., and Wolf, J. B. (2014). The evolution of genomic imprinting: theories, predictions and empirical tests. Heredity, 113(2):119–128. Queller, D. C. (2003). Theory of genomic imprinting conflict in social insects. Bmc Evolutionary Biology, 3:art. no.–15. Team, R.-C. (2013). R: A language and environment for statistical computing. Wang, X. and Clark, A. G. (2014). Using next-generation RNA sequencing to identify imprinted genes. Heredity. Weber, F., Fukino, K., Villalona-Calero, M., and Eng, C. (2005). Limitations of Single-Strand Conformation Polymorphism Analysis As a High-Throughput Method for the Detection of EGFR Mutations in the Clinical Setting. Journal of Clinical Oncology, 23(24):5847–5848. Weiner, S. A. and Toth, A. L. (2012). Epigenetics in social insects: a new direction for understanding the evolution of castes. Genetics research international, 2012:609810. Welch, M. and Lister, R. (2014). Epigenomics and the control of fate, form and function in social insects. Current Opinion in Insect Science, 1:31–38. Yan, H., Simola, D. F., Bonasio, R., Liebig, J., Berger, S. L., and Reinberg, D. (2014). Eusocial insects as emerging models for behavioural epigenetics. Nature Reviews Genetics, 15(10):677–688.

9/9 PeerJ PrePrints | http://dx.doi.org/10.7287/peerj.preprints.878v1 | CC-BY 4.0 Open Access | rec: 7 Mar 2015, publ: 7 Mar 2015